Forty-two patients were enrolled from 12 institutions in Japan between May 2008 and November 2010. All patients were eligible. Patients’ characteristics are shown in Table 1. The Consolidated Standards of Reporting Trials (CONSORT) diagram is summarized in Fig. 1.
Neoadjuvant chemotherapy was started in 42 patients and discontinued in three because of progressive disease (n = 1), AE (n = 1), or both (n = 1). The remaining 39 patients completed three courses of induction chemotherapy without requiring any dose reduction. Thus, the chemotherapy completion rate was 92.9 % (39/42). There was one case with grade 4 toxicity of neutropenia (Table 2). Grade 3 toxicity was observed in seven patients. There were no chemotherapy-induced deaths. According to the mRECIST criteria, 14 patients achieved partial remission (PR), 19 had stable disease (SD), and seven had progressive disease (PD). Two patients were not evaluable because they had no radiologically evaluable lesions. Response and disease-control rates were 33.3 % [95 % confidence interval (CI) 19.6–49.6 %] and 78.6 % (95 % CI 63.2–89.7 %), respectively.
Of the 39 patients who completed chemotherapy, six did not proceed to EPP because of progressive disease (n = 4), AE (n = 1), or protocol violation (n = 1). EPP was started in 33 patients and was completed in 30 (14 right sided, 16 left sided). These 30 patients achieved MCR with reconstruction of the diaphragm (n = 2), pericardium (n = 2), or both (n = 26). EPP was abandoned in three patients because of extensive chest wall invasion (n = 2) or transmural pericardial invasion (n = 1). Accordingly, the MCR rate was 71.4 % (30/42). Median operation time and blood loss in completed EPP cases (n = 30) were 437 (range 335–655) min and 1461 (range 390–4530) g, respectively.
There were four TRDs, and all occurred within 84 days after EPP (Table 3). One patient died of bronchopleural fistula, empyema, and acute respiratory disease syndrome (ARDS) on postoperative day (POD) 30. Another patient developed cardiac herniation and hemothorax on POD 23 and died on POD 80. Two patients died of ARDS on PODs 61 and 72.
Postoperative complications are summarized in Table 4. Grade 3 or greater postoperative complications occurred in 21 of 33 patients (63.6 %). Arrhythmia occurred in 11 (33.3 %) but was not critical in all cases. Bronchopleural fistula developed in three patients who underwent right-sided EPP.
Of the 30 patients with MCR, 11 did not proceed to RT because of AEs (n = 7), tumor relapse (n = 2), protocol violation (n = 1), or personal decision (n = 1). Of the 19 patients who started RT, two discontinued because of AE (n = 1) or refusal (n = 1). Accordingly, TMT was completed in 17 (40.5 %, n = 42) patients. Grade 3 toxicity related to RT occurred in four patients (Table 5). There were no patients with grade ≥4 AEs or grade ≥3 radiation pneumonitis or interstitial pneumonia.
Preoperative histological diagnosis of epithelioid MPM was revised to biphasic MPM postoperatively for two out of 30 patients who completed EPP. In the other three patients with a preoperative diagnosis of undetermined MPM, two were diagnosed as having sarcomatoid and one as having epithelioid MPM.
Survival, RFS, and recurrence
The median follow-up period was 19.9 months after registration. MST and the 2-year survival rate for all registered patients (n = 42) were 19.9 (95 % CI 14.2–27.3 %) and 42.9 % (95 % CI 27.8–57.1 %) months, respectively (Fig. 2a). MST and the 2-year survival rate of patients who completed EPP (n = 30) were 22.7 months and 50.0 %, respectively, and those for non-EPP patients (n = 12) were 17.1 months and 25.0 %, respectively (Fig. 3).
Among the 30 patients who completed EPP with MCR, 19 relapsed postoperatively, with a median RFS of 11.0 (95 % CI 2.2–31.5 %) months; RFS rates 1 and 2 years after surgery were 57.6 and 37.0 %, respectively (Fig. 2b). Relapse occurred at ipsilateral pleural effusion/chest wall (n = 8), mediastinum/pericardial effusion (n = 7), contralateral lung parenchyma (n = 2), ascites (n = 3), lymph nodes (n = 3), and unknown (n = 1). Of 17 patients who completed TMT, relapse occurred in 11 at ipsilateral pleural effusion/chest wall (n = 5), mediastinum/pericardial effusion (n = 2), contralateral lung parenchyma (n = 2), and ascites (n = 2). Significantly longer survival was observed in patients with a preoperative diagnosis of epithelioid histology compared with nonepithelioid histology (MST 27.3 vs. 13.6 months, P = 0.0013), and there were no statistically significant differences in survival according to other preoperative variables, such as age, gender, side, histology, clinical stage, and radiological response to chemotherapy.
There was no statistically significant survival difference correlated with pathological T/N factors and pathological stage. Significantly longer survival was observed for patients who completed TMT (n = 17) in comparison with patients who completed EPP but not TMT (n = 13) (MST 39.4 vs. 11.4 months, P = 0.0243) (Fig. 4).